An improved Genetic Algorithm of Bi-level Coding for Flexible Job Shop Scheduling Problems

The current study presents an improved genetic algorithm(GA) for the flexible job shop scheduling problem (FJSP). The coding is divided into working sequence level and machine level and two effective crossover operators and mutation operators are designed for the generation and reduce the disruptive effects of genetic operators. The algorithm is tested on instances of 10 working sequences and 10 machines. Computational results show that the proposed GA was successfully and efficiently applied to the FJSP. The results were compared with other approaches, such as traditional GA and GA with neural network. Compared to traditional genetic algorithm, the proposed approach yields significant improvement in solution quality.